Two studies on non-verbal processes were conducted in 1974-1975. In one, white undergraduates taught a short lesson to a 10 year old child who was either white or black. The undergraduates were told on a random basis that the child was either "bright" or "dull" or were given no information. Then, the tutor-child interaction was secretly video- taped, and later analyzed for non-verbal behavior by blind raters. It was found that, across induced expectancies, tutors exhibited more positive non-verbal behavior to white children than to black. In addition, tutors were more positive (smiles, eyegaze, etc.) to bright whites than control or dull whites. However, expectation of intelligence had no effect on tutor behavior toward black children. In another study, a teacher on video-tape simulated one of three styles of non-verbal behavior: positive, neutral or negative. Children watched her and then took a test on the lesson. Results indicated that different styles of non-verbal behavior had no effect on student performance. It was hypothesized that perhaps a face-to-face interaction is necessary for non-verbal behavior to affect performance. An experiment to test that hypothesis is being planned for the coming year.